3. Results
Lung field flow values are given in
Table 2 and
Table 3. In 50% (17/34) of cases, the flow in the upper field was greater than the average upper field perfusion of 21.9%. In 38.2% (13/34) of patients, the flow ranged from 23.3% to 28.5%, meeting the criteria for high (23.3–24.8%) and almost full or full use (25–28.5%) of flow reserves [
9,
10,
11,
12,
13,
14,
15].
In 53% (18/34) of patients, the transverse dimension of the RA was equal to or greater than the average value of 4 cm. In 17.6% (6/34) of patients, the transverse dimension of the RA was in the range of 4.3–6 cm, indicating its widening (>4.3 cm) [
16]. In 47% (16/34) of patients, the longitudinal dimension of the RA was greater than the mean value of 5.2 cm. In 32.3% (11/34) of patients, the longitudinal dimension of the RA was within the range indicating its widening (5.7–6.4 cm; reference value < 5.4 cm) [
16]. In 50% (17/34) of patients, the transverse dimension of the RV was greater than the mean value of 4.3 cm. However, in 64.7% (22/34) of patients, the transverse dimension of the RV was within the range indicating its widening (4.2–5.2 cm; reference value < 4.1 cm) [
17]. The longitudinal dimension of the RV was greater than the average value of 8.5 cm in 44.1% (15/34) of patients [
17]. In this subgroup, the longitudinal dimension of the RV was in the range of 8.6–11 cm, indicative of its widening (reference value < 8.3 cm) [
17].
Furthermore, in the analysis of the surgical side, we found significant differences in the longitudinal dimensions of the RA and RV depending on side of operation. The patients with the remaining left lung had a significantly larger longitudinal dimension of the RA compared to the patients with the remaining right lung (5.7 cm vs. 4.7 cm;
p = 0.0001). There was also a significantly larger longitudinal dimension of the RV in patients with the remaining right lung compared to patients with the remaining left lung (9 cm vs. 8 cm;
p = 0.025). In 32.35% (11/34) of patients, the diameter of the PA behind its main division (PA I) was greater than the mean value of 2.4 cm. In 26.5% (9/34) of patients, PA I met the criteria for extension (>2.51 cm) [
18]. In the other two patients with PA I > 2.4 cm, the diameter almost reached 2.5 cm. In 35.3% (12/34) of patients, the diameter at the site of the first segmental branch (PA II) was greater than the mean value of 2.2 cm. In 23.5% (8/34) of patients PA II was within the range indicating its widening (2.4–2.8 cm; reference value > 2.3 cm). In 44.1% (15/34) of patients, the diameter of the PA at the level of the main trunk (PA III) was greater than the mean value of 2.8 cm; 26.5% (9/34) of patients met the criteria for dilatation (3–3.5 cm; reference value > 2.95 cm). Furthermore, in 17.6% (6/34) of patients, the diameter of the PA had already reached the limiting dimension of 2.9 cm [
19]. The individual dimensions are given in
Table 4 and
Table 5.
The mean pulmonary perfusion in the upper field was 21.9%. Among patients with >21.9% perfusion in the upper field, significantly lower pulmonary flow values were recorded in the lower field (29% vs. 33%,
p = 0.031) and in the lower and middle fields combined (73.5% vs. 79.6%,
p < 0.001) compared to patients with <21.9% perfusion (
Figure 1).
In the group with pulmonary flow in the upper field > 21.9%, the transverse dimension of the RA (mean 4.1 cm vs. 3.8 cm,
p = 0.003) and all dimensions of the PA were significantly greater (PA I: 2.5 cm vs. 2.3 cm,
p = 0.001; PA II: 2.3 cm vs. 2.1 cm,
p = 0.008; PA III: 3 cm vs. 2.6 cm,
p < 0.001;
Figure 2).
The transverse dimension of the RA was 4 cm on average. Among patients with a >4 cm transverse dimension of the RA, significantly higher perfusion values were recorded in the upper field (26.1% vs. 19.8%,
p < 0.001) but significantly lower perfusion values in the lower field (28% vs. 32.4%,
p = 0.006) and in the lower and middle fields (72.4% vs. 78.5%,
p < 0.001) compared to patients with a <4 cm transverse dimension of the RA (
Figure 3).
In this comparison, a significantly greater longitudinal dimension of the RA (5.5 cm vs. 5 cm,
p = 0.046) and transverse dimension of the RV (4.6 cm vs. 4.2 cm,
p = 0.007) were also noted (
Figure 4). Furthermore, all PA diameters (PA I–III) were significantly greater among patients with a >4 cm transverse dimension of the RA (PA I: 2.7 cm vs. 2.3 cm,
p < 0.001; PA II: 2.4 cm vs. 2.1 cm,
p < 0.001; PA III: 3.1 cm vs. 2.7 cm,
p < 0.001) compared to patients with a <4 cm transverse dimension of the RA (
Figure 5). Differences in the longitudinal dimension of the RV based on the transverse dimension of the RA (>4 cm vs. <4 cm) were not significant (8.4 cm vs. 8.8 cm;
p = 0.46).
The transverse dimension of the RV was 4.3 cm on average. Among patients with a >4.3 cm transverse dimension of the RV, a significantly greater transverse dimension of the RA was recorded (4.2 cm vs. 3.8 cm;
p = 0.029) than among patients with a <4.3 cm transverse dimension of the RV (
Figure 6). In this subgroup, the differences in terms of upper field perfusion (22.07% vs. 21.7%;
p = 0.70), PA I (2.5 cm vs. 2.4 cm;
p = 0.099), PA II (2.3 cm vs. 2.1 cm;
p = 0.12), and PA III (2.9 cm vs. 2.7 cm;
p = 0.09), and the longitudinal dimension of the RV (8.1 cm vs. 8.9 cm;
p = 0.09) did not reach significance.
The longitudinal dimension of the RV was 8.5 cm on average. There were no significant differences in the transverse dimension of the RV (4.2 cm vs. 4.5 cm, p = 0.12) or any of the dimensions of the RA (transverse: 3.9 cm vs. 4 cm, p = 0.31; longitudinal: 5.2 cm vs. 5.1 cm, p = 0.6) or PA (PA I: 2.4 cm vs. 2.5 cm, p = 0.11; PA II: 2.1 cm vs. 2.3 cm, p = 0.19; PA III: 2.7 cm vs. 2.9 cm, p = 0.08) based on the longitudinal dimension of the RV being < 8.5 cm or >8.5 cm. However, we noted a significant difference related to weight (67.5 kg vs. 79.5 kg, p = 0.03) and BMI (23.7 vs. 27.6, p = 0.03).
The comparison of the diameters of the PA was performed based on the diameter of PA I, which was 2.4 cm on average. Among patients with a PA I > 2.4 cm, significantly greater pulmonary flow was observed in the upper field (26.1% vs. 19.8%,
p < 0.001) and significantly lower pulmonary perfusion in the lower field (28% vs. 32.4%,
p = 0.006) and in the lower and middle fields (72.4% vs. 78.5%,
p < 0.001;
Figure 7).
Among patients with a PA I > 2.4 cm, the transverse and longitudinal dimensions of the RA were significantly greater (4.5 cm vs. 3.7 cm,
p < 0.001 and 5.5 cm vs. 5 cm,
p = 0.046, respectively), as was the transverse dimension of the RV (4.6 cm vs. 4.2 cm,
p = 0.007) compared to patients with a PA I < 2.4 cm (
Figure 8). In this subgroup (PA I > 2.4 cm), significantly greater PA II (2.4 cm vs. 2.1 cm,
p < 0.001) and PA III (3.1 cm vs. 2.7 cm,
p < 0.001) measurements were also observed (
Figure 9), but we did not find significant differences in the longitudinal dimension of the RV (8.4 cm vs. 8.8 cm,
p = 0.46).
We also found numerous positive and negative correlations between the above-mentioned features, which are outlined in
Table 6 and
Table 7.
We found positive correlations between pulmonary perfusion in the upper field, PA diameter, and transverse dimensions of the RA and RV consistent with the state of and changes in the pulmonary circulation. The negative correlations we identified concerned opposing perfusion features in individual lung fields and the related dimensions of the PA. The above relationships excluded the longitudinal dimension of the RV, which correlated with BMI and weight.
The absence of features of PH or the presence of a single feature of PH was observed in 8 patients. A set of at least five coexisting features of PH was found in another 8 patients. However, the majority of patients (
n = 18) had two to four features of PH. A summary of the analyzed features in relation to literature standards depending on the time from pneumonectomy is presented in
Table 8,
Table 9,
Table 10 and
Table 11.
An analysis of the relationship between the results and published norms among observations made up to 1 year after pneumonectomy revealed that 60% of cases were within the accepted limits or exceeded the norms for only a single feature (0 to 1), whereas 40% of cases had three to eight coexisting features of PH.
The observations made 1–5 years after pneumonectomy revealed that 37.5% of cases exceeded the norms for one to two features of PH, and 62.5% exceeded the norms for three to eight coexisting features of PH (
Table 9).
In the period 5–10 years from pneumonectomy, 75% of cases either fell within the accepted limits or exceeded the norms for zero to two features, and in 25% of cases a syndrome of three coexisting features of PH was observed (
Table 10).
The observations made >10 years after pneumonectomy revealed four to seven coexisting features of PH in all patients. However, there were no physiological relationships between the analyzed features of the pulmonary circulatory system or with individual PH features. The above trend does not include the longitudinal dimension of the RV, as widening occurred in only one case (
Table 11).
4. Discussion
Stabilization of respiratory function after pneumonectomy is accompanied by an increase in the flow in the remaining lung, remodeling of the circulatory system, including both the PA and right heart cavities, and subsequent development of features of PH syndrome [
1,
5,
6,
7,
20,
21,
22,
23,
24]. This is attributed to the use of flow reserves in the remaining lung to stabilize the PA pressure, but elevated systolic pressure in the RV with features of mild PH have been reported on the second day after removal of an entire lung [
25]. Increased flow in the one-lung system is accompanied by an increase in pressure in the PA and leads to dilatation of the right heart cavities and the remaining branch of the PA as a cascade of postoperative events [
1,
5,
6,
7,
8,
20,
26,
27]. Lung circulation changes involve the use of previously inactive areas within the pulmonary capillaries, which is particularly evident in the upper fields of the remaining lung [
9,
10,
11,
23]. Perfusion of the single upper field should not exceed 10% of the total lung perfusion in a two-lung system [
9,
10,
11,
12]. Previous studies noted 21.35% perfusion after pneumonectomy [
11]. In the current study, the average perfusion level of the upper field was 21.9%, which is consistent with previous observations. The importance of the level of upper field perfusion is that increasing the flow from <10% in the two-lung system to ≥21.9% in the one-lung system intensively uses the lung flow reserves; after exceeding 25%, it practically exhausts them as a mechanism for stabilizing the pulmonary flow and counteracting PH [
28]. This situation concerned 13 patients in the study group with submaximal or maximal flow in the range of 23.3–28.5%, particularly 9 patients with flow in the range of 25–28.5%. According to Chandra, increased perfusion in the upper field is associated with an unfavorable prognosis, and the limiting value is 20% of flow [
13]. Another phenomenon that was observed when the perfusion of the upper field increased was a proportional decrease in perfusion in the other fields of the remaining lung as one common flow area. This observation is important in regard to changes in the ventilation–perfusion (V/Q) ratio and the effectiveness of the blood oxygenation process [
29].
The mean diameter of the PA at the point of division into the right and left branches (i.e., PA I) was 2.4 cm. One-third of the patients in the study group had a PA I > 2.4 cm, and more than one-fourth had a PA I greater than the reference value of 2.51 cm [
18]. The mean PA diameter at the level of the main trunk (i.e., PA III) was 2.8 cm. Nearly half of the patients in the study group had a PA III > 2.8 cm, and more than one-fourth had a PA III greater than the reference value of 29.5 cm [
19,
30]. This observation has an additional, and at the same time unfavorable, importance; according to Żyłkowska, an enlarged PA diameter is recognized as an independent risk factor for death in patients with PH [
31]. No data were found in the literature regarding the PA diameter at the site of the first segmental branch (i.e., PA II). The obtained average of 2.2 cm may be the basis for future comparisons and analyses.
The mean transverse and longitudinal dimensions of the RA in the study group were 4 cm and 5.2 cm, respectively. According to Maceira, the reference values for mean RA dimensions for the European population are 4.3 cm and 5.4 cm, respectively [
16]. More than half (53%) of the patients in the study group had a transverse dimension of the RA ≥ 4 cm, and nearly half (47%) of the patients had a longitudinal dimension of the RA > 5.2 cm. In Maceira’s data, nearly one-sixth (17.6%) of the RA transverse dimensions and nearly one-third (32.3%) of the RA longitudinal dimensions were greater than the cited reference dimensions. A relationship between RA changes and PH was described by Alenezi [
32]. Moreover, enlargement, and even dysfunction, of the RA was reported in a group of patients with PH [
33,
34], and a positive correlation between increased systolic pressure in the PA and RA dilatation in patients with PH was observed by Cioffi [
35]. According to the literature, enlargement of the dimensions of the RA and PA is associated with the development of PH and with potential worsening of the prognosis [
31,
32,
33,
35,
36,
37,
38,
39,
40,
41]. Other negative prognostic factors in pulmonary hypertension include signs of right heart failure, frequent syncope, elevated pulmonary vascular resistance, elevated right atrial pressure, reduced tricuspid annular plane systolic excursion (TAPSE), tricuspid regurgitation, pericardial effusion, reduced right ventricular ejection fraction, and decreased right ventricular septal and global longitudinal strain [
42,
43,
44,
45].
Furthermore, RV dilatation after pneumonectomy has been reported in the literature [
6,
7,
20,
26,
46,
47]. Kowalewski demonstrated an increase in the dimensions and deterioration of the systolic function of the RV 2 days after pneumonectomy, which was confirmed in subsequent studies [
5,
7,
25,
48]. The particular value of analyzing the dimensions of the RV lies in the combination of its dilated morphology and impairment of its systolic function, though the efficiency of the RV was not examined in the current study [
7,
8,
25,
38,
47,
48,
49,
50]. The relationship between increased flow in the remaining lung, increased pressure and size in the PA, and dilatation of the right chambers of the heart was also described by Foroulis and Venuta [
1,
5,
6,
7,
8,
20,
26]. Interestingly, the ratio of TAPSE and systolic pulmonary arterial pressure (sPAP) was introduced as a marker of the coupling of pulmonary circulation and right ventricle [
51]. Moreover Tello et al. demonstrated the prognostic value of this parameter in PH patients. Lower TAPSE/sPAP ratio was associated with worse overall survival [
52]. Sonaglioni et al. also demonstrated the predictive value of TAPSE/sPAP ratio in patients with idiopathic pulmonary fibrosis [
53]. Unfortunately, we did not measure this parameter in our study. The mean transverse and longitudinal dimensions of the RV in the current study were 4.3 cm and 8.5 cm, respectively. In the NORRE study, the reference dimensions of the RV were 3.4 cm and 6.8 cm, respectively, but the physiological transverse dimension could reach 4.1 cm and the longitudinal dimension up to 8.3 cm [
54]. Błażejewski in his analysis reported maximum normal RV basal and longitudinal diameters in accordance with American and European Echocardiographic Guidelines (4.1 cm and 8.3 cm, respectively) [
17]. Almost all patients were characterized by a larger transverse and longitudinal dimension of the RV than the cited standards. However, differences in imaging techniques should be taken into account when interpreting these results. This selective interpretation also applies to the current study and constitutes its limitation.
The relationships observed in the pulmonary circulation are closely related, with a statistical correlation between the level of pulmonary flow, the dimensions of the PA, and most of the dimensions of the right heart cavities (see
Table 6 and
Table 7). Perfusion in the upper field significantly positively correlated with all sizes of PA I, PA II, and PA III and significantly negatively correlated with the perfusion in the lower field and perfusion in the combined lower and middle fields. In the group of patients with PA I > 2.4 cm, significantly greater transverse and longitudinal dimensions of the RA, transverse dimension of the RV, and other dimensions of the PA (i.e., PA II and PA III) were demonstrated (
Figure 8 and
Figure 9). The reported relationships are based on earlier publications by Żyłkowska, Ratanawatkul, and Oganesian [
15,
30,
31]. Among patients with a transverse dimension of the RA > 4 cm, significantly higher perfusion was observed in the upper field of the remaining lung with significantly lower flow in the lower field and in the combined lower and middle fields. In this group of patients, significant enlargement of the longitudinal dimension of the RA, transverse dimension of the RV, and all three diameters of the PA were also noted. The association of PH with RA dilatation and increased systolic pressure in the PA in patients with PH was reported by Cioffi, and the association of an increase in PA diameter, systolic pressure, and RV systolic functional status in patients with PH was reported by Tonelli [
35,
36]. The transverse dimension of the RV is part of these relationships. For patients with a transverse dimension of the RV > 4.3 cm, significantly greater transverse dimensions of the RA were found. In addition, the value of the transverse dimension of the RV significantly positively correlated with all diameters of the PA. There are many studies in the literature on PH regarding RV augmentation [
6,
7,
20,
26,
45,
55]. Fisher reported a positive correlation of systolic pressure in the PA and vascular resistance with RA and RV enlargement in patients with PH [
38]. Enlargement of the RV diameter, muscle wall area, and outflow tract in a group of patients with PH was also described by Swift et al. [
56]. However, we did not identify a statistical correlation between the longitudinal dimension of the RA and pulmonary perfusion in the upper field or PA I. Another type of relationship was observed for the longitudinal dimension of the RV, significant positive correlations with weight and BMI. The lack of other relationships for the longitudinal dimension of the RV may indicate an asymmetric influence of the two dimensions of the RV and, in some cases, the predominant influence of the transverse dimension in the process of RV dilatation.
The majority of studies have described changes in the immediate period after lung removal and in the first 5 years after pneumonectomy [
7,
20,
25,
26,
48]. The condition of patients > 5 years after pneumonectomy was documented by Deslauries [
5], but longer term observations have been rare. The literature indicates that the process of PH development does not proceed via the same dynamics in all patients. It seems that the characteristics of this process were most accurately presented by Potaris, who demonstrated an increase in systolic pressure in the PA and the development of PH in nearly 40% of patients within the first year after pneumonectomy [
57]. The results obtained in the present study were similar to previous observations. The flow reserves in the lung were used or exceeded in almost 40% of patients. Dilatation of the right heart chambers occurred in 17.6–64.7% of patients depending on the size examined, as well as in 23.5–26.5% of patients when analyzing PA diameter. PH symptoms were present in 40% of patients already in the first year of follow-up, 62.5% of patients 1–5 years after pneumonectomy, 25% of patients 5–10 years after pneumonectomy, and all patients who were followed >10 years after the procedure. Moreover, a severe set of PH features was detected in nearly a quarter of cases.
The results of the present study indicate that, in the adopted model of increased perfusion in the upper field of a single lung, dilatation of the RA, RV, and PA develops at a different rate during the first 10 years after the procedure. However, the small group size limits our conclusions. A similar situation and limitations will apply to the global dimension for cases requiring removal of the entire lung for oncological reasons [
4].